CN114125745A - MQTT protocol power control and QoS mechanism selection method - Google Patents
MQTT protocol power control and QoS mechanism selection method Download PDFInfo
- Publication number
- CN114125745A CN114125745A CN202111375386.0A CN202111375386A CN114125745A CN 114125745 A CN114125745 A CN 114125745A CN 202111375386 A CN202111375386 A CN 202111375386A CN 114125745 A CN114125745 A CN 114125745A
- Authority
- CN
- China
- Prior art keywords
- packet
- transmission
- data packet
- power control
- qos
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 230000007246 mechanism Effects 0.000 title claims abstract description 194
- 238000010187 selection method Methods 0.000 title claims description 20
- 230000005540 biological transmission Effects 0.000 claims abstract description 140
- 238000005265 energy consumption Methods 0.000 claims abstract description 91
- 238000000034 method Methods 0.000 claims abstract description 41
- 230000008569 process Effects 0.000 claims description 21
- 230000009471 action Effects 0.000 claims description 15
- 238000004422 calculation algorithm Methods 0.000 claims description 12
- 238000004364 calculation method Methods 0.000 claims description 5
- 238000010276 construction Methods 0.000 claims description 3
- 230000004044 response Effects 0.000 claims description 3
- 238000005457 optimization Methods 0.000 abstract description 22
- 238000004891 communication Methods 0.000 abstract description 12
- 238000012545 processing Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 3
- 238000012790 confirmation Methods 0.000 description 2
- 230000008878 coupling Effects 0.000 description 2
- 238000010168 coupling process Methods 0.000 description 2
- 238000005859 coupling reaction Methods 0.000 description 2
- 230000007423 decrease Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000004927 fusion Effects 0.000 description 2
- 238000004088 simulation Methods 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000011217 control strategy Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000006855 networking Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000011664 signaling Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/30—Services specially adapted for particular environments, situations or purposes
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16Y—INFORMATION AND COMMUNICATION TECHNOLOGY SPECIALLY ADAPTED FOR THE INTERNET OF THINGS [IoT]
- G16Y10/00—Economic sectors
- G16Y10/35—Utilities, e.g. electricity, gas or water
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W28/00—Network traffic management; Network resource management
- H04W28/02—Traffic management, e.g. flow control or congestion control
- H04W28/06—Optimizing the usage of the radio link, e.g. header compression, information sizing, discarding information
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W28/00—Network traffic management; Network resource management
- H04W28/16—Central resource management; Negotiation of resources or communication parameters, e.g. negotiating bandwidth or QoS [Quality of Service]
- H04W28/24—Negotiating SLA [Service Level Agreement]; Negotiating QoS [Quality of Service]
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W52/00—Power management, e.g. TPC [Transmission Power Control], power saving or power classes
- H04W52/02—Power saving arrangements
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Quality & Reliability (AREA)
- Business, Economics & Management (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- Economics (AREA)
- General Business, Economics & Management (AREA)
- Computing Systems (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The application relates to the field of power distribution internet of things data transmission, and particularly discloses a method for selecting a power control and QoS mechanism of an MQTT protocol, which comprises the steps of firstly obtaining the type of the MQTT power control and QoS mechanism; and then, obtaining a compromise transmission scheme determined based on the weighted sum of the transmission packet loss rate and the energy consumption in the protocol power and QoS mechanism through a preset MAB model based on historical transmission information and local information of the equipment side, and performing data transmission. Therefore, through the MAB model, joint optimization of power control and QoS mechanism selection of the MQTT protocol under incomplete information and dynamic compromise of packet loss rate and energy consumption are achieved, and accordingly differentiated QoS communication requirements of power distribution Internet of things services are met.
Description
Technical Field
The application relates to the technical field of data transmission of the Internet of things, in particular to a method for power control and QoS mechanism selection of an MQTT protocol.
Background
The power distribution internet of things realizes comprehensive sensing, data fusion and intelligent application of the power distribution network through interconnection and intercommunication among power distribution equipment. With the rapid development of the internet of things technology in the power distribution network, due to the limited computing power and battery capacity, the existing power distribution equipment is difficult to meet the Quality of Service (QoS) communication requirements of the power distribution internet of things services, such as low packet loss rate, low energy consumption and the like. Therefore, how to use limited resources is a very difficult problem to achieve QoS communication requirements with low energy consumption while ensuring low packet loss rate.
Disclosure of Invention
The application provides a method for power control and QoS mechanism selection of an MQTT protocol, which aims to solve the problems that in the prior art, power distribution equipment is difficult to meet the service quality communication requirements of power distribution Internet of things services with low packet loss rate, low energy consumption and the like, namely the QoS communication requirements of low energy consumption can not be realized while the low packet loss rate can not be ensured, compromise between the packet loss rate and the low energy consumption can not be realized, and thus an optimal comprehensive scheme can be selected.
The above object of the present application is achieved by the following technical solutions:
the embodiment of the application provides a method for selecting a power control and QoS mechanism of an MQTT protocol, which comprises the following steps:
acquiring MQTT power control and QoS mechanism types;
and obtaining a compromise transmission scheme determined based on the weighted sum of the transmission packet loss rate and the energy consumption in the protocol power and QoS mechanism through a preset MAB model based on historical transmission information and local information of the equipment side, and performing data transmission.
Further, the preset MAB model construction process includes:
defining preset power distribution Internet of things equipment as a player in the MAB model;
defining all possible choices of MQTT power control and QoS mechanisms as swing arms in the MAB model;
defining the negative number of the weighted sum of packet loss rate and energy consumption as the return in the MAB model in the data transmission process;
defining MQTT protocol power control and QoS mechanism selection as actions in an MAB model;
defining the basis for MQTT protocol power control and QoS mechanism selection as the strategy in the MAB model.
Further, the obtaining, by presetting the MAB model, a compromise transmission scheme determined based on a weighted sum of transmission packet loss rate and energy consumption in the protocol power and QoS mechanism based on historical transmission information and device side local information, and performing data transmission includes:
initializing all index variables to be zero, randomly generating a first preset value, and determining an optimal MQTT protocol power control and QoS mechanism selection strategy;
when the current data packet is transmitted, judging and selecting an MQTT protocol power control and QoS mechanism based on preference estimated values of the equipment to transmission power and the QoS mechanism, and updating the selected times of the estimated values and the mechanism;
and continuing the mechanism selection and transmission of the next data packet based on the updated estimation value and the selected times of the mechanism until all the data packets are transmitted.
Further, when the current data packet is transmitted, based on the preference estimation value of the device for the transmission power and the QoS mechanism, judging and selecting the MQTT protocol power control and QoS mechanism, and updating the estimation value and the number of times the mechanism is selected, including:
when the current data packet is transmitted, acquiring preference estimated values of the equipment to transmission power and a QoS mechanism;
based on the preference estimation value and a preset judgment formula, selecting a mechanism with the maximum preference estimation value or randomly selecting a transmission mechanism from all transmission mechanisms;
the estimate is updated as well as the number of times the mechanism is selected.
Further, the QoS mechanisms include QoS0 mechanism, QoS1 mechanism, and QoS2 mechanism;
the QoS0 mechanisms include: the sending end sends the data packet to the receiving end and deletes the stored data packet after the sending action is finished;
the QoS1 mechanisms include: the sending end sends a data packet to the receiving end, and deletes the stored data packet after receiving the feedback data packet sent by the receiving end in preset time; or when the feedback data packet sent by the receiving end is not received within the preset time, the data packet is continuously sent to the receiving end until the feedback data packet sent by the receiving end is received within the preset time after the data packet is sent latest, and the stored data packet is deleted;
the QoS2 mechanisms include: the sending end sends a data packet to the receiving end, deletes the discarded data packet after receiving a first feedback data packet sent by the receiving end within preset time, sends a second feedback data packet to the receiving end, so that the receiving end sends a third feedback data packet to the sending end after receiving the second feedback information, and ends data transmission after receiving the third feedback data packet; and the sending end also resends the data packet to the receiving end after not receiving the first feedback data packet sent by the receiving end within the preset time.
Further, the packet loss ratio calculation of each QoS mechanism includes: determining the packet loss rate of data packet transmission based on the channel gain, the signal-to-noise ratio threshold value required by successful data transmission and the current signal-to-noise ratio during data transmission;
the calculation of the energy consumption of each of the QoS mechanisms comprises: the total energy consumption is determined based on the number of transmission processes of the data transmission mechanism and the deduplication energy consumption.
Further, the energy consumption of the QoS0 mechanism is as follows:
wherein,
gi,j,1denotes the channel gain, H, of the n-th retransmission of the j-th packet transmitting the i-th large packeti,j,nDenotes the channel frequency response at the nth transmission of the ith large packet, the jth small packet, n0S represents the data amount of each small packet, B represents the bandwidth, and p (i) is the transmission power of the ith large packet;
the energy consumption of the QoS1 mechanism is that the total power consumption is the total energy consumption of the common transmission power consumption to increase the total energy consumption of the multiple receiving data packets, and specifically:
wherein, the total energy consumption for repeatedly receiving the data packet is as follows:
E0deduplication energy consumption for a single packet;
the common transmission power consumption is as follows:
wherein,
a retransmission indication variable under the QoS1 mechanism for the nth retransmission of the jth small packet in the ith large packet,
a feedback indicator variable p representing the nth feedback of the feedback data packet of the ith large packet and the jth small packet under the QoS1 mechanismbackIndicating the return power of the data packet, gi,j,n,backIndicating the channel state, N, at the time of packet returni,jIndicates the total number of transmission times, S, of the ith large packet and the jth small packetBACKIndicating the size of the feedback information;
the energy consumption of the QoS2 mechanism is as follows:
wherein,
and
retransmission indicator variable and return indicator variable for the data packet and the first feedback data packet under the QoS2 mechanism respectively,
and
retransmission indicating variables and return indicating variables of the second feedback data packet and the third feedback data packet under the QoS2 mechanism respectively;
Further, the MAB model is:
where m is the mechanism in QoS, k is power control,
negative number, x, of weighted sum of packet loss rate and energy consumptioni,m,kTo indicate variables for actions, C3And C4And the MQTT QoS mechanism selection and the power control variable constraint of the power distribution internet of things are represented.
Further, the basis for selecting the MQTT protocol power control and QoS mechanism is defined as a policy in the MAB model, wherein the policy is determined based on epsilon-greedy algorithm.
The technical scheme provided by the embodiment of the application can have the following beneficial effects:
in the technical scheme provided by the embodiment of the application, MQTT power control and QoS mechanism types are firstly obtained; and then, obtaining a compromise transmission scheme determined based on the weighted sum of the transmission packet loss rate and the energy consumption in the protocol power and QoS mechanism through a preset MAB model based on historical transmission information and local information of the equipment side, and performing data transmission. Therefore, through the MAB model, joint optimization of power control and QoS mechanism selection of the MQTT protocol under incomplete information and dynamic compromise of packet loss rate and energy consumption are achieved, and accordingly differentiated QoS communication requirements of power distribution Internet of things services are met.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the application.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and together with the description, serve to explain the principles of the application.
Fig. 1 is a schematic view of an application scenario of a method for selecting a power control and QoS mechanism of an MQTT protocol according to an embodiment of the present application;
fig. 2 is a schematic flowchart of a MQTT protocol power control and QoS mechanism selection method according to an embodiment of the present application;
FIG. 3 is a diagram illustrating the relationship between optimization objectives and the number of big packet transmissions;
fig. 4 is a schematic diagram of a variation trend of packet loss rate and energy consumption along with a weight V in the MQTT protocol power control and QoS mechanism selection method provided in the embodiment of the present application.
Detailed Description
Reference will now be made in detail to the exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, like numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present application. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present application, as detailed in the appended claims.
The power distribution internet of things realizes comprehensive sensing, data fusion and intelligent application of the power distribution network through interconnection and intercommunication among power distribution equipment. With the rapid development of the internet of things technology in the power distribution network, due to the limited computing power and battery capacity, the existing power distribution equipment is difficult to meet the Quality of Service (QoS) communication requirements of the power distribution internet of things services, such as low packet loss rate, low energy consumption and the like. Therefore, how to use limited resources is a very difficult problem to achieve QoS communication requirements with low energy consumption while ensuring low packet loss rate.
Message Queue Telemetry Transport (MQTT) is an internet of things Transport protocol based on a publish/subscribe mechanism, has the characteristics of simplicity, light weight, high bandwidth utilization rate and the like, and is an effective method for solving the problems. In particular, the proxy server sends messages to subscribers who subscribe to a particular topic and uses smaller packet headers to reduce signaling overhead. In addition, it provides three flexible QoS mechanisms to satisfy reliable transmission of different task data, namely "distribute QoS0 at most once", "distribute QoS1 at least once", and "distribute QoS2 once". For the QoS0 level, the device transmits the message to the proxy server at most once, and when the device sends the message, the message temporarily stored on the device is deleted, and the sending end does not care whether the receiving end receives the message, so that the device has a higher packet loss rate, but the time delay and the energy consumption are relatively low. For the QoS1 level, successful transmission of a message from the device to the receiver occurs at least once, and the device deletes the buffered message only when confirming that the receiver receives the message, or repeatedly sends the message to the receiver with successful transmission, thereby ensuring that the message arrives accurately. However, multiple transmissions may result in the receiving end receiving repeated messages, and the receiving end needs to consume more energy to perform the repeated message packet deduplication operation. For the QoS2 level, the device transmits a message to the receiving end, ensuring that the message arrives only once and not repeatedly by repeatedly confirming whether the message arrives at the receiving end. But multiple retransmissions and backhauls result in relatively high transmission power consumption. In summary, the MQTT QoS mechanism selection may have a large impact on energy consumption and packet loss, and energy consumption may be effectively reduced through power control. However, power control in turn affects packet loss rate performance by affecting the signal-to-noise ratio. Therefore, in order to realize dynamic compromise between packet loss rate and energy consumption, dynamic joint optimization of MQTT protocol power control and QoS mechanism selection is required.
However, joint optimization of MQTT protocol power control with QoS mechanism selection still presents some key technical challenges. First, the conventional MQTT protocol power control and QoS mechanism selection needs to be based on system global information, i.e., time-varying context information, including channel state information, bandwidth information, etc. But in an actual implementation environment, this information is unknown to the device. Second, the optimization problem is an NP-hard problem due to the coupling of MQTT protocol power control with QoS mechanism selection. Moreover, this exacerbates the complexity of the optimization problem. Specifically, the MQTT QoS mechanism selection policy affects the power control policy, which also affects the MQTT QoS mechanism selection policy. Thirdly, how to realize dynamic compromise between packet loss rate and energy consumption is a difficult problem. In particular, it is not practical to achieve both the lowest energy consumption and the lowest packet loss rate, such as when the QoS2 mechanism is selected, although the low packet loss rate is achieved, multiple transmissions of task data may result in higher energy consumption for transmission. If the QoS0 mechanism is selected, the packet loss rate increases although the transmission power consumption is low and the deduplication power consumption is not required. Therefore, how to realize dynamic compromise of packet loss rate and energy consumption through dynamic joint optimization of MQTT protocol power control and QoS mechanism selection under incomplete information to meet the differentiated QoS communication requirements of the power distribution internet of things services still remains a problem to be solved.
In order to solve the problems, the application provides a method for selecting a power control and QoS mechanism of an MQTT protocol, and particularly relates to a method for selecting a power distribution Internet of things, so that compromise selection is made for packet loss rate and energy consumption of ten thousand networking devices in a transmission process, and therefore the QoS communication requirement of low energy consumption is met while the first packet loss rate is ensured. Specific embodiments are illustrated in detail by the following examples.
Examples
First, an application scenario of the MQTT protocol power control and QoS mechanism selection method provided by the present application is introduced, fig. 1 is a schematic view of an application scenario of the MQTT protocol power control and QoS mechanism selection method provided by the present application embodiment, as shown in fig. 1:
the MQTT protocol power control and QoS mechanism selection method is applied to the field of power distribution Internet of things, and dynamic compromise between packet loss rate and energy consumption is achieved through a power distribution Internet of things oriented MQTT protocol power control and QoS mechanism selection combined optimization scheme. An application scenario of the power control and QoS mechanism selection method of the power distribution Internet of things oriented MQTT protocol is shown in fig. 1, and the power control and QoS mechanism selection method of the power distribution Internet of things comprises N power distribution Internet of things devices, 1 base station equipped with an MQTT gateway, 1 Internet of things cloud platform and a power distribution Internet of things application control center. The specific flow of communication is as follows. Firstly, the power distribution internet of things equipment sends a communication request to the gateway based on an MQTT protocol, and the gateway forwards related configuration information to the power distribution internet of things equipment. And secondly, the power distribution internet of things equipment sends the task data to an internet of things cloud platform for processing based on an MQTT protocol. And finally, the Internet of things cloud platform returns the task processing result to the power distribution Internet of things application control center. The present application is directed to solving the second process, the device send data process.
Fig. 2 is a schematic flow chart of a MQTT protocol power control and QoS mechanism selection method provided in the embodiment of the present application, and as shown in fig. 2, the MQTT protocol power control and QoS mechanism selection method provided in the embodiment of the present application includes the following steps:
s101, obtaining MQTT power control and QoS mechanism types.
In practical applications, it is assumed that there are a total of I large packets of task data on the nth device. Wherein each large packet contains J small packets. Supposing that when the transmission of each big packet task data is started, the equipment side performs MQTT protocol power control and QoS mechanism selection, and in the transmission process of each big packet task data, the MQTT protocol power control and QoS mechanism selection strategy is kept unchanged until the transmission of the big packet task data is finished. Defining m (i) ∈ {0,1,2} as MQTT protocol QoS mechanism selection variable of ith big packet, where m (i) ═ 0 denotes selecting QoS0 mechanism, m (i) ═ 1 denotes selecting QoS1 mechanism, and m (i) ═ 2 denotes selecting QoS2 mechanism. We define p (i) as the transmission power of the ith large packet, where p (i) e { p }1,p2,…,pk,…,pKThere are K power control modes. Three QoS mechanisms, packet loss rate and energy consumption models of the MQTT protocol are specifically described below.
First, it is the mechanism that distributes QoS0 once at most: the sending end sends a PUBLISH data packet containing a message to the receiving end, and each data packet is sent only once, without concern of packet loss of the receiving end, and without retransmission process and information confirmation return process.
Second is the mechanism of distributing QoS1 at least once: the QoS1 mechanism is abbreviated, each data packet is guaranteed to be transmitted at least once successfully, and if the receiving end feedback confirmation message is not received within a period of time, the data packet is retransmitted until the receiving end successfully receives the data packet and the transmitting end successfully receives the feedback message of the receiving end. The specific process is as follows: firstly, a sending end sends a PUBLISH packet with data to a receiving end, and the PUBLISH packet is stored locally. Secondly, after receiving the PUBLISH packet, the receiving end sends a PUBLISH packet containing the information header but without data content to the sending end. During this period, after sending the PUBLISH packet, the sending end waits for the feedback of the receiving end, and the upper limit of the waiting time is described as t0, if the PUBLISH packet is received in t0, the sending of the message is completed, and if the PUBLISH packet is not received, that is, the PUBLISH packet fails to be sent back, the sending end continues to send the PUBLISH packet. In the above process, the receiving end will take the received PUBLISH data packet each time as a new data packet and return the PUBLISH data packet, and whenever the return of the PUBLISH data packet fails, the transmitting end will retransmit the PUBLISH data packet, which results in the receiving end receiving repeated data packets, and a 'deduplication' action is required to remove the received heavy data packet, thereby increasing deduplication energy consumption of the receiving end.
The third is to distribute the QoS2 mechanism only once: abbreviated as the "QoS 2 mechanism," each packet is sent only once successfully. The QoS2 mechanism ensures that each data packet is successfully sent only once through at least two backhaul procedures, and the specific flow is as follows: firstly, a sending end sends a PUBLISH data packet to a receiving end, and the PUBLISH data packet is stored locally. Secondly, after receiving the PUBLISH data packet, the receiving end locally stores the information header of the PUBLISH data packet, and replies to the sending end with a PUBLISH data packet without data content, and if the receiving end receives a data packet with the information header consistent with the information header stored before, the receiving end considers that the data packet is a repeated message and discards the repeated message. Then, when the sending end receives the PUBREC data packet, it discards the initially stored PUBLISH data packet, stores the PUBREC data packet at the same time, and replies to a PUBREL data packet without data content at the receiving end. And finally, after the receiving end receives the PUBREL data packet, replying a PUBCOMP data packet without data content to the sending end. And when the transmitting end successfully receives the PUBCOMP data packet, the transmission process is considered to be finished. Similar to the QoS1 mechanism, the sender waits for a feedback time after sending a data packet, the upper limit of which is t0, and if no message is received within t0, the data packet is retransmitted. Different from QoS1, the QoS2 transmission mechanism includes at least two backhauls, and the receiving end stores the PUBLISH information header and discards the PUBLISH data packet with the same information header, so that the receiving end is guaranteed not to receive repeated messages, and thus, no duplicate removal energy consumption exists in the QoS2 mechanism, however, two backhauls increase transmission delay, that is, the QoS2 mechanism avoids duplicate removal energy consumption at the cost of increasing the possibility of delay.
And S102, obtaining a compromise transmission scheme based on the weighting sum of transmission packet loss rate and energy consumption in the protocol power and QoS mechanism based on historical transmission information and equipment side local information through a preset MAB model, and performing data transmission.
It should be noted that, when performing power control and QoS mechanism selection, the two most important factors are packet loss rate and power consumption of data transmission. For the packet loss rate, the QoS0 mechanism has a packet loss phenomenon, the other two mechanisms do not have packet loss, and for the energy consumption QoS0 mechanism, the retransmission and return process does not exist, and the duplicate removal energy consumption is not contained; the QoS1 mechanism has retransmission and backhaul processes, but the backhaul process is only successful once, and has deduplication energy consumption; under the QoS2 mechanism, at least two backhauls exist in each data packet, and the data packet does not contain the deduplication energy consumption. The details will be described below separately.
Specifically, in the present application, it is assumed that the channel state is unchanged during any packet task data transmission process, but the channel state randomly changes between any packet data transmission processes. Therefore, the channel gain of the nth retransmission of the jth packet transmitting the ith large packet is defined as
Wherein Hi,j,nWhich represents the channel frequency response at the nth transmission of the ith large packet and the jth small packet, and n0 is the noise power. Since the QoS0 mechanism message is sent only once, n in the QoS0 mechanism is 1; the QoS1 mechanism and QoS2 mechanism allow for retransmissions, and each retransmission is marked as a packet transmission process, i.e., a channel between any retransmission processesThe state changes.
(1) For the QoS0 mechanism. Definition of
And a packet loss indicating variable under the QoS0 mechanism for the jth small packet in the ith big packet. Wherein,
the judgment expression shows that the jth small packet in the ith big packet has packet loss, otherwise, the jth small packet has no packet loss, and is as follows
Wherein G isthA signal-to-noise ratio threshold required for successful task data transmission. That is, when the current signal-to-noise ratio is lower than the threshold, packet loss occurs, otherwise, no packet loss occurs. The packet loss rate of the ith large packet is
Wherein,
is the total number of lost packets of the ith large packet, and can be expressed as
Since the QoS0 mechanism does not have deduplication power consumption, the total power consumption is the transmission power consumption. The energy consumption of the ith large packet is
Where S denotes a data amount (bit) of each packet, and B denotes a bandwidth.
(2) For the QoS1 mechanism. Since the QoS1 mechanism certainly ensures that the task data transmission is successful, no packet loss occurs. Therefore, under the QoS1 mechanism, the ith is largePacket loss rate of packets is
Definition of
Retransmission indicator variable under QoS1 mechanism for nth retransmission of PUBLISH data packet of jth small packet in ith large packet. Wherein,
the nth transmission of the PUBLISH data packet representing the ith big packet and the jth small packet fails, otherwise, the transmission succeeds, and the judgment expression is as follows
Wherein p isbackIndicating the return power of the data packet, gi,j,n,backIndicating the channel state of the data packet during its return. Definition of
The feedback indicator variable is the feedback indicator variable of the feedback data packet, which is the feedback data packet, of the last big packet and the last small packet of the ith big packet, and is transmitted back for the nth time under the QoS1 mechanism. Wherein,
the n-th retransmission of the backup data packet of the jth packet representing the ith big packet fails, and the judgment expression is as follows:
the transmission energy consumption of the jth packet of the ith large packet is expressed as:
wherein N isi,jTo representTotal number of transmissions of ith large packet and jth small packet, SBACKIndicating the size of the heartbeat packet. The first item of the formula represents transmission energy consumption under the condition of transmission failure of the PUBLISH data packet, the second item of the formula represents retransmission energy consumption of the PUBLISH data packet and postback energy consumption of the PUBACK data packet caused by postback failure of the PUBACK data packet, the third item and the fourth item of the formula represent transmission success of the PUBLISH data packet and transmission energy consumption under the condition of postback success of the PUBACK packet, and when the transmission success of the PUBLISH data packet and the transmission success of the PUBACK data packet, the transmission is completed.
Under the QoS1 mechanism, the receiving end may receive duplicate PUBLISH packets and need to perform self-deduplication. The number of repeatedly transmitted PUBLISH packets is represented by the number of times that the transmission of the PUBLISH data packets is successful and the return of the PUBREC data packets is failed, and the energy consumption of the repeatedly received PUBLISH packets is represented as
Wherein E is0The deduplication power consumption for a single PUBLISH packet. Total energy consumption is expressed as
(3) For the QoS2 mechanism. Similar to the QoS1 mechanism, the QoS2 mechanism also ensures that task data is successfully transmitted without packet loss. The packet loss rate of the ith large packet is
The QoS2 mechanism includes the following processes: transmitting a PUBLISH data packet; the PUBREC data packet is a first feedback data packet and is transmitted back; transmitting a PUBREL data packet, namely a second feedback data packet; the PUBCOMP packet, i.e. the third feedback packet, is returned to four processes. The PUBREL data packet is transmitted only after the PUBREC data packet is successfully transmitted back. For convenience of description, the first two transmission processes are collectively referred to as a first transmission process under the QoS2 scheme, i.e., a first predetermined transmission process, and the second two transmission processes are collectively referred to as a second transmission process under the QoS2 scheme, i.e., a second predetermined transmission process.
During the first transmission we define
And
the retransmission indicator variable and the return indicator variable are respectively the retransmission indicator variable and the return indicator variable of the PUBLISH data packet and the PUBREC data packet under the QoS2 mechanism. In the second transmission we define
And
the retransmission indication variable and the return indication variable are respectively the retransmission indication variable and the return indication variable of the PUBREL data packet and the PUBCOMP data packet under the QoS2 mechanism. The judgment method of the above indication variables is similar to that of the QoS 1. The transmission energy consumption of the first transmission process of the jth packet of the ith large packet under the QoS2 mechanism is as follows:
the transmission energy consumption of the second transmission process of the jth packet of the ith large packet is as follows:
the total energy consumption of the ith large packet is:
further, after determining the calculation mode of the packet loss rate and the energy consumption, performing subsequent mechanism selection through model construction.
Specifically, a distributed joint optimization problem is firstly established, namely, the power distribution internet of things equipment minimizes the weighted sum of packet loss rate and energy consumption by making an optimal MQTT protocol power control and QoS mechanism selection strategy according to local information and historical information. Thus, the joint optimization problem is modeled as
Wherein V is a non-negative weight parameter for balancing the importance of energy consumption and packet loss rate. C1And C2And the power distribution internet of things MQTT QoS mechanism selection and power control variable constraint are represented, namely, each large packet can only select one MQTT protocol power control and QoS mechanism selection strategy.
The optimization problem is then transformed into a MAB problem and scientifically and rationally designed for players, rockers, rewards, actions and strategies as described below.
A player: the body executing the strategy and generating the action can continuously update the strategy by learning the return value fed back by the history. Herein, we define the power distribution internet of things device as a player. Rocker arm: the player can make candidates for action, how many candidates are, i.e. how many rockers. In this context, we define all possible options for MQTT protocol power control and QoS mechanism selection as a rocker arm, i.e. a
A total of 3K rocker arms. And (3) returning: feedback information received after each round player takes action. In this context, we define the reward as packet lossNegative of weighted sum of rate and energy consumption, i.e.
The actions are as follows: the player depresses the rocker arm. In this context, we define MQTT protocol power control and QoS mechanism selection as actions. And we define xi,m,kFor action indicating variables, the expression is:
wherein x isi,m,k1 denotes that the transmission power of the ith large packet is pkAnd selects the QoSm mechanism, otherwise xi,m,k0. Strategy: the basis of the selection action. In this context, we define the basis for MQTT QoS mechanism selection and power control as a policy, and devise a policy based on epsilon-greedy algorithm.
From the modeling described above, the original problem can be converted into:
wherein, C3And C4C corresponding to original problem1And C2The method indicates that only one MQTT QoS mechanism and power control strategy can be selected for each big packet, and the MQTT protocol power control and QoS mechanism selection strategy is not changed during the data processing of each big packet.
In some specific implementation processes, after determining a problem-solving model, the method for selecting the MQTT protocol power control and QoS mechanism includes the following specific selection processes:
(1) and (5) an initialization phase. First all the indicator variables are initialized to zero, and we assume that the device selects each mode once for the first 3K large packets to be transmitted. And then generating a random number mu epsilon (0,1), and selecting an optimal MQTT protocol power control and QoS mechanism selection strategy according to a corresponding formula.
(2) And (5) a learning stage. Firstly, when the ith big data packet is transmitted, the device transmits power p to the transmissionkAnd a preference estimation for QoSm mechanisms of
According to the following
Carrying out MQTT protocol power control and QoS mechanism selection when mu is>When epsilon, selecting the mechanism with the maximum preference estimation value, otherwise, randomly selecting the mechanism from the 3K transmission mechanisms; finally according to
ri,m,k＝ri-1,m,k+xi,m,kAnd updating the estimated value and the selected times of the mechanism, and entering the transmission process of the next big data packet.
(3) And (5) terminating the phase. When I +1, learning is terminated.
The MQTT protocol power control and QoS mechanism selection method provided by the application is a combined optimization scheme for power distribution Internet of things for MQTT protocol power control and QoS mechanism selection. Firstly, constructing a joint optimization problem of MQTT protocol power control and QoS mechanism selection under various actual constraint conditions, wherein the optimization target is to minimize the weighted sum of packet loss rate and energy consumption. Secondly, the MQTT protocol power control and QoS mechanism selection combined optimization problem is constructed into a Multi-arm Bandit (MAB) problem. And finally, based on the historical information and the local information of the equipment side, the online learning capacity of the epsilon-greedy algorithm is utilized to realize MQTT protocol power control and QoS mechanism selection.
Therefore, firstly, joint optimization of MQTT protocol power control and QoS mechanism selection under incomplete information is realized, an original optimization problem is constructed into an MAB problem, and the problem of coupling of optimization variables is solved. Secondly, historical information is observed and learned by using an epsilon-greedy algorithm, and optimal MQTT protocol power control and QoS mechanism selection are realized under the condition that only local side information is known. And secondly, realizing dynamic compromise between packet loss rate and energy consumption. The present invention sets the optimization objective to minimize the weighted sum of packet loss rate and energy consumption. The weight parameter reflects the importance of the packet loss rate and the energy consumption. Therefore, by adjusting the weight parameter, dynamic compromise between packet loss rate and energy consumption can be realized, and the differentiated QoS (quality of service) requirements of the power distribution internet of things service are met.
To illustrate the superiority of the method of the present invention, simulation experiments were conducted. Wherein p is1＝35mW，p2＝25mW，p back30 mW. The total number of the transmitted big data packets is set to be 800, each big data packet is divided into 10 small data packets for transmission, the channel state is unchanged in the transmission process of any small data packet, and the channel state randomly changes between the transmission processes of any small data packet. Fig. 3 is a relationship between an optimization objective and a number of large packet transmissions. It can be seen from FIG. 3 that the curve shows the QoS1/p at the first 300 large packet transmissions1The pattern performed best, all curves trended down after 300 large packet transmissions, QoS0/p1The decline is faster and the performance is better. The reason is that when the 300-time and 800-time large packet transmission is performed, the channel quality is improved, the packet loss rate is obviously reduced under the QoS0 mechanism, and the QoS1 and the QoS2 have less weight influence on energy consumption due to the fact that the packet loss rate is 0, and the weighting and reduction trend of the packet loss rate and the energy consumption is not obvious. The numerical result shows that the proposed algorithm is compared with single selection QOS0/p in weighted sum of packet loss rate and energy consumption1And QoS2/p2The decrease was 11.58% and 49.10%, respectively. This is because the proposed algorithm can learn the channel state after channel changes, from QoS1/p1Is switched to QoS0/p1。
Fig. 4 shows the variation trend of packet loss rate and energy consumption with the weight V in the transmission mode of the algorithm proposed in the present application. It can be observed from fig. 4 that as the weight V increases, the energy consumption shows a downward trend, and the packet loss rate shows an upward trend. Because the influence of energy consumption is gradually increased along with the increase of V, and the influence of packet loss rate is gradually weakened, the proposed algorithm can measure the optimization of packet loss and energy consumption according to the weight value so as to adapt to the transmission under the differentiated QoS requirement of the power distribution internet of things.
Therefore, extensive performance evaluation is completed, and compared with the existing algorithm, the algorithm has outstanding performances in packet loss rate, transmission delay and the like through a large amount of simulation verification. Specifically, the invention proves the superiority of the algorithm by changing different parameters, such as the number of data packets and the loss threshold of the data packets, and has good popularization and application prospects.
It is understood that the same or similar parts in the above embodiments may be mutually referred to, and the same or similar parts in other embodiments may be referred to for the content which is not described in detail in some embodiments.
It should be noted that, in the description of the present application, the terms "first", "second", etc. are used for descriptive purposes only and are not to be construed as indicating or implying relative importance. Further, in the description of the present application, the meaning of "a plurality" means at least two unless otherwise specified.
Any process or method descriptions in flow charts or otherwise described herein may be understood as representing modules, segments, or portions of code which include one or more executable instructions for implementing specific logical functions or steps of the process, and the scope of the preferred embodiments of the present application includes other implementations in which functions may be executed out of order from that shown or discussed, including substantially concurrently or in reverse order, depending on the functionality involved, as would be understood by those reasonably skilled in the art of the present application.
It should be understood that portions of the present application may be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, the various steps or methods may be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, any one or combination of the following techniques, which are known in the art, may be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application specific integrated circuit having an appropriate combinational logic gate circuit, a Programmable Gate Array (PGA), a Field Programmable Gate Array (FPGA), or the like.
It will be understood by those skilled in the art that all or part of the steps carried by the method for implementing the above embodiments may be implemented by hardware related to instructions of a program, which may be stored in a computer readable storage medium, and when the program is executed, the program includes one or a combination of the steps of the method embodiments.
In addition, functional units in the embodiments of the present application may be integrated into one processing module, or each unit may exist alone physically, or two or more units are integrated into one module. The integrated module can be realized in a hardware mode, and can also be realized in a software functional module mode. The integrated module, if implemented in the form of a software functional module and sold or used as a stand-alone product, may also be stored in a computer readable storage medium.
The storage medium mentioned above may be a read-only memory, a magnetic or optical disk, etc.
In the description herein, reference to the description of the term "one embodiment," "some embodiments," "an example," "a specific example," or "some examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the application. In this specification, the schematic representations of the terms used above do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
Although embodiments of the present application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present application, and that variations, modifications, substitutions and alterations may be made to the above embodiments by those of ordinary skill in the art within the scope of the present application.
Claims (9)
1. A method for selecting MQTT protocol power control and QoS mechanism is characterized by comprising the following steps:
acquiring MQTT power control and QoS mechanism types;
and obtaining a compromise transmission scheme determined based on the weighted sum of the transmission packet loss rate and the energy consumption in the protocol power and QoS mechanism through a preset MAB model based on historical transmission information and local information of the equipment side, and performing data transmission.
2. The MQTT protocol power control and QoS mechanism selection method according to claim 1, wherein the construction process of the preset MAB model comprises:
defining preset power distribution Internet of things equipment as a player in the MAB model;
defining all possible choices of MQTT power control and QoS mechanisms as swing arms in the MAB model;
defining the negative number of the weighted sum of packet loss rate and energy consumption as the return in the MAB model in the data transmission process;
defining MQTT protocol power control and QoS mechanism selection as actions in an MAB model;
defining the basis for MQTT protocol power control and QoS mechanism selection as the strategy in the MAB model.
3. The MQTT protocol power control and QoS mechanism selection method according to claim 1, wherein obtaining, in the protocol power and QoS mechanism, a compromise transmission scheme determined based on a weighted sum of transmission packet loss rate and energy consumption through a preset MAB model based on historical transmission information and device side local information, and performing data transmission, comprises:
initializing all index variables to be zero, randomly generating a first preset value, and determining an optimal MQTT protocol power control and QoS mechanism selection strategy;
when the current data packet is transmitted, judging and selecting an MQTT protocol power control and QoS mechanism based on preference estimated values of the equipment to transmission power and the QoS mechanism, and updating the selected times of the estimated values and the mechanism;
and continuing the mechanism selection and transmission of the next data packet based on the updated estimation value and the selected times of the mechanism until all the data packets are transmitted.
4. The MQTT protocol power control and QoS mechanism selection method according to claim 3, wherein the determining to select the MQTT protocol power control and QoS mechanism and updating the estimation value and the number of times the mechanism is selected based on the preference estimation value of the device for the transmission power and QoS mechanism at the time of the current packet transmission comprises:
when the current data packet is transmitted, acquiring preference estimated values of the equipment to transmission power and a QoS mechanism;
based on the preference estimation value and a preset judgment formula, selecting a mechanism with the maximum preference estimation value or randomly selecting a transmission mechanism from all transmission mechanisms;
the estimate is updated as well as the number of times the mechanism is selected.
5. The MQTT protocol power control and QoS mechanism selection method according to claim 1, wherein the QoS mechanisms include a QoS0 mechanism, a QoS1 mechanism, and a QoS2 mechanism;
the QoS0 mechanisms include: the sending end sends the data packet to the receiving end and deletes the stored data packet after the sending action is finished;
the QoS1 mechanisms include: the sending end sends a data packet to the receiving end, and deletes the stored data packet after receiving the feedback data packet sent by the receiving end in preset time; or when the feedback data packet sent by the receiving end is not received within the preset time, the data packet is continuously sent to the receiving end until the feedback data packet sent by the receiving end is received within the preset time after the data packet is sent latest, and the stored data packet is deleted;
the QoS2 mechanisms include: the sending end sends a data packet to the receiving end, deletes the discarded data packet after receiving a first feedback data packet sent by the receiving end within preset time, sends a second feedback data packet to the receiving end, so that the receiving end sends a third feedback data packet to the sending end after receiving the second feedback information, and ends data transmission after receiving the third feedback data packet; and the sending end also resends the data packet to the receiving end after not receiving the first feedback data packet sent by the receiving end within the preset time.
6. The MQTT protocol power control and QoS mechanism selection method of claim 5, wherein,
the packet loss rate calculation of each QoS mechanism includes: determining the packet loss rate of data packet transmission based on the channel gain, the signal-to-noise ratio threshold value required by successful data transmission and the current signal-to-noise ratio during data transmission;
the calculation of the energy consumption of each of the QoS mechanisms comprises: the total energy consumption is determined based on the number of transmission processes of the data transmission mechanism and the deduplication energy consumption.
7. The MQTT protocol power control and QoS mechanism selection method of claim 6, wherein,
the energy consumption of the QoS0 mechanism is as follows:
wherein,
gi,j,1denotes the channel gain, H, of the n-th retransmission of the j-th packet transmitting the i-th large packeti,j,nRepresenting the channel frequency response of the ith big packet at the nth transmission of the jth small packet, wherein n0 is noise power, S represents the data volume of each small packet, B represents bandwidth, and p (i) is the transmission power of the ith big packet;
the energy consumption of the QoS1 mechanism is that the total power consumption is the total energy consumption of the common transmission power consumption to increase the total energy consumption of the multiple receiving data packets, and specifically:
wherein, the total energy consumption for repeatedly receiving the data packet is as follows:
E0deduplication energy consumption for a single packet;
the common transmission power consumption is as follows:
wherein,
a retransmission indication variable under the QoS1 mechanism for the nth retransmission of the jth small packet in the ith large packet,
a feedback indicator variable p representing the nth feedback of the feedback data packet of the ith large packet and the jth small packet under the QoS1 mechanismbackIndicating the return power of the data packet, gi,j,n,backIndicating the channel state, N, at the time of packet returni,jIndicates the total number of transmission times, S, of the ith large packet and the jth small packetBACKIndicating the size of the feedback information;
the energy consumption of the QoS2 mechanism is as follows:
wherein,
and
retransmission indicator variable and return indicator variable for the data packet and the first feedback data packet under the QoS2 mechanism respectively,
and
retransmission indicating variables and return indicating variables of the second feedback data packet and the third feedback data packet under the QoS2 mechanism respectively;
8. The MQTT protocol power control and QoS mechanism selection method of claim 5, wherein the MAB model is as follows:
P2:
where m is the mechanism in QoS, k is power control,
negative number, x, of weighted sum of packet loss rate and energy consumptioni,m,kTo indicate variables for actions, C3And C4And the MQTT QoS mechanism selection and the power control variable constraint of the power distribution internet of things are represented.
9. The MQTT protocol power control and QoS mechanism selection method according to claim 2, wherein the basis for MQTT protocol power control and QoS mechanism selection is defined as a policy in an MAB model, wherein the policy is determined based on epsilon-greedy algorithm.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111375386.0A CN114125745A (en) | 2021-11-19 | 2021-11-19 | MQTT protocol power control and QoS mechanism selection method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111375386.0A CN114125745A (en) | 2021-11-19 | 2021-11-19 | MQTT protocol power control and QoS mechanism selection method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114125745A true CN114125745A (en) | 2022-03-01 |
Family
ID=80396823
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111375386.0A Pending CN114125745A (en) | 2021-11-19 | 2021-11-19 | MQTT protocol power control and QoS mechanism selection method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114125745A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113472843A (en) * | 2021-05-24 | 2021-10-01 | 国网山东省电力公司电力科学研究院 | Greedy algorithm based MQTT protocol QoS mechanism selection method |
| CN116248582A (en) * | 2022-12-30 | 2023-06-09 | 深圳市乐科智控科技有限公司 | Internet of things communication method, electronic equipment and Internet of things system |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113132490A (en) * | 2021-04-26 | 2021-07-16 | 华北电力大学 | MQTT protocol QoS mechanism selection scheme based on reinforcement learning |
| CN113438053A (en) * | 2021-06-23 | 2021-09-24 | 中宇联云计算服务(上海)有限公司 | Packet loss compensation method, system and equipment based on cloud network fusion technology |
| CN113472843A (en) * | 2021-05-24 | 2021-10-01 | 国网山东省电力公司电力科学研究院 | Greedy algorithm based MQTT protocol QoS mechanism selection method |
-
2021
- 2021-11-19 CN CN202111375386.0A patent/CN114125745A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113132490A (en) * | 2021-04-26 | 2021-07-16 | 华北电力大学 | MQTT protocol QoS mechanism selection scheme based on reinforcement learning |
| CN113472843A (en) * | 2021-05-24 | 2021-10-01 | 国网山东省电力公司电力科学研究院 | Greedy algorithm based MQTT protocol QoS mechanism selection method |
| CN113438053A (en) * | 2021-06-23 | 2021-09-24 | 中宇联云计算服务(上海)有限公司 | Packet loss compensation method, system and equipment based on cloud network fusion technology |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113472843A (en) * | 2021-05-24 | 2021-10-01 | 国网山东省电力公司电力科学研究院 | Greedy algorithm based MQTT protocol QoS mechanism selection method |
| CN116248582A (en) * | 2022-12-30 | 2023-06-09 | 深圳市乐科智控科技有限公司 | Internet of things communication method, electronic equipment and Internet of things system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111629380B (en) | Dynamic resource allocation method for high concurrency multi-service industrial 5G network | |
| CN114125745A (en) | MQTT protocol power control and QoS mechanism selection method | |
| WO2021227508A1 (en) | Deep reinforcement learning-based industrial 5g dynamic multi-priority multi-access method | |
| CN112770353B (en) | Method and device for training congestion control model and method and device for controlling congestion | |
| CN108684046B (en) | Random learning-based access network service function chain deployment method | |
| CN113132490A (en) | MQTT protocol QoS mechanism selection scheme based on reinforcement learning | |
| CN113055489A (en) | Implementation method of satellite-ground converged network resource allocation strategy based on Q learning | |
| CN113315716B (en) | Training method and equipment of congestion control model and congestion control method and equipment | |
| CN113966596B (en) | Method and apparatus for data traffic routing | |
| CN101296152A (en) | Data scheduling method and system of equity linked network | |
| WO2024113974A1 (en) | Computing power network routing allocation method and apparatus, electronic device, and storage medium | |
| CN115189908B (en) | Random attack survivability evaluation method based on network digital twin | |
| CN116566838A (en) | Internet of vehicles task unloading and content caching method with cooperative blockchain and edge calculation | |
| CN111211988A (en) | Data transmission method and system for distributed machine learning | |
| CN115189910B (en) | Intentional attack survivability evaluation method based on network digital twin | |
| CN116489708A (en) | Meta universe oriented cloud edge end collaborative mobile edge computing task unloading method | |
| CN116489104A (en) | Traffic scheduling method and system based on dynamic priority | |
| CN113472843A (en) | Greedy algorithm based MQTT protocol QoS mechanism selection method | |
| CN113014499B (en) | Data transmission method and device, electronic equipment and storage medium | |
| Bhattacharyya et al. | QFlow: A learning approach to high QoE video streaming at the wireless edge | |
| EP4024212A1 (en) | Method for scheduling interference workloads on edge network resources | |
| CN113596138B (en) | Heterogeneous information center network cache allocation method based on deep reinforcement learning | |
| TWI792784B (en) | Method and system for federated reinforcement learning based offloading optimization in edge computing | |
| CN116668439A (en) | Edge computing task unloading method based on deep reinforcement learning | |
| CN115756873A (en) | Mobile edge computing unloading method and platform based on federal reinforcement learning |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20220301 |